Decentralized Diagnosis in Heavy Duty Vehicles

Fault diagnosis is important for automotive vehicles, due to economic reasons, such as efficient repair and fault prevention, and legislative reasons, mainly safety and pollution. Embedded systems in vehicles includes a large number of electronic control units, that is connected to each other via an electronic network. Many of the current diagnostic systems use pre-compiled diagnostic tests, and fault-logs that stores the results from the tests. To improve the diagnostic system, fault localization in addition to the existing fault detection is wanted. Since there are limitations in processing power, memory, and network capacity, an algorithm is searched for that uses stated diagnoses in the control units to find the diagnoses for the complete system. Such an algorithm is presented and exemplified in the article. The embedded system used in a Scania heavy duty vehicle, has been used as a case study to find limitations in the embedded system, and realistic requirements on the algorithm. Copyright c © 2004 J. Biteus, M. Jensen, & M. Nyberg.